IT partners are grappling with commercial questions about AI and tempered expectations about its adoption, according to Tech Research Asia research director Trevor Clarke.
Clarke will lay out the business realities at CRN Pipeline in August, where he will reveal the results of the TRA-CRN survey of CxOs and channel firms.
The results include which AI-related services CxOs are considering engaging third parties for help with, which phase of AI adoption they are in and what steps they have taken to prepare for AI. Clarke will also share his own insights about the IT channel-AI business pathway, gleaned from extensive work with IT channel firms, vendors and end user businesses.
“The appetite [from CxOs for partners to help them with AI] is absolutely there,” Clarke said.
But the commercial model for GenAI is a work in progress for many IT channel firms, Clarke said.
“For most of the channel, the primary concern is figuring out how to generate revenue from AI, and we're seeing this issue come up frequently.”
“There are significant questions: How do we seize the opportunity? Should we establish a separate AI practice or integrate AI capabilities into our existing practices? And crucially, how do we monetise it?”
“There's so many ideas that you can come up with. You can literally come up with huge long lists of possible use cases,” Clark noted.
“But the challenge that we’re seeing for a lot of partners is how do they go from a proof of concept for one of those ideas and then turning that into a full deployment with the managed services attached to it and all the other attached services and products out there?”
"We often encounter partners and vendors who remain uncertain about the precise customer journey in AI,” Clarke said. “They struggle to identify where it starts and then how are they actually going through that process.”
Off-the-shelf AI tools such as GenAI assistants may save companies from developing their own solutions. But without clear workflows, tasks and objectives, AI initiatives risk “devolving into generic attempts at boosting productivity”, Clarke commented.
Saying “‘Hey, this is going to help productivity, let's give people access to it’ – that sort of thing doesn't necessarily work.”
“You really need that rigor and framework, like you do with a typical IT project, to make sure that it works. The off-the-shelf, embedded stuff – you've got to really put the effort into making sure that works.”
AI might also require changes to work habits. “Unless you change the way people actually work then you've just deployed something that is adding to the mix of stuff that people use on a daily basis.”
IT firms have seen this before. “You know, there's some parallels or similarities here back to social enterprise networks and things like Chatter and Yammer, which turned into Slack and Teams and that sort of thing. It took a very long time before we all started actually using those effectively in the right way.”
“So, there is an element of change management here and changing culture, which, as we've all learned over many, many years, takes a long time to actually truly embed.”
This necessitates a consultative approach rather than a typical IT sale, in Clarke’s opinion.
“That’s a big change,” he said. “Do a lot of partners actually have salespeople that can truly engage in business consulting, not just IT consulting? That’s where a lot of people we know are investing and trying to recruit really good people.”
Where wins and losses are determined
“Choosing the right proof of concepts and use cases is really where wins and losses are being seen when it comes to AI business at the moment,” Clarke said.
“There really aren’t any commodified use cases yet,” he said. “There may be some small things like using a GenAI service to produce blog posts for your content marketing, sure. That's not really going to make you that much money as a partner.”
“We’re in that exploratory phase of finding those use cases that we can productise and take to customers, so that we can then start to have some repeatable processes and refine that so that we’re really good at delivering that and getting financial benefits as a result.”
“The challenge is so many of the use cases are unique to each organisation, each workflow in that organisation.”
He points to work by large consulting firms to tailor AI for specific functions in specific industries.
“It’s HR in banking and specific roles in that industry. It’s ‘How do we help with insurance claims processing?’ It’s ‘How do we do government forms for the citizen experience with filling out forms and trying to get funding for things?’”
“I think what we’ll start to see is people come out and say ‘We can really help you if you're in manufacturing with this specific challenge that you have, or this specific workflow, by applying what we've learned with AI for this.’”
“It’s very much around applying AI to existing workflows as opposed to trying to create new stuff with AI. That’s the opportunity.”
“The new stuff – as in, ‘We're going to build stuff, we're going to build our own IP’ – is a much longer-term opportunity and it is a very different kind of conversation.”
“It's not that there's not opportunity there, there is absolutely is. But most channel partners will not have the capability to do that unless they're in a niche already and that's their core business.”
“It's very hard to build a new practice for building AI for customers. It’s very hard to invest because you're going to compete on remuneration for getting great people who are already being targeted to work for other companies who have probably a lot more money than you do.”
“It’s not a horizontal opportunity”
Looking ahead, Clarke expects more granular categorisation of AI, reflecting its application across different domains, for industrial processes to other industry-specific uses.
“It’s not a horizontal opportunity. This is very much going after existing customers in vertical ways with a consulting-led approach.”
“It’s kind of amplifying all the trends around verticalisation and being more business-led and knowing your customers’ business more. All that stuff has been amplified as a result of the AI conversation at the moment and that's where the opportunity is heading in most places.”
Trevor Clarke will reveal more insights in the sessions “Evolution of the IT landscape in 2024” and “Improving cloud, security & AI profits” at CRN Pipeline on August 13 and 14. See the Pipeline agenda and register your interest in attending.
-1.jpg&w=100&c=1&s=0)
.png&w=100&c=1&s=0)
